Acupuncture for Sleep Fragmentation in Older Adults: A Single-Arm Pilot Study

Change in 7-day mean wake after sleep onset from baseline to Week 6

Change equals the Week 6 mean minus the baseline mean. Wake after sleep onset (WASO) is the total minutes scored awake between sleep onset and final awakening for each valid night. Sleep is measured with Actiwatch 2 (Philips Respironics) worn on the nondominant wrist, recorded in 60-second epochs, and scored with Philips Actiware version 6.0.9 using sleep diaries. Each timepoint is the mean across all valid nights in the 7-day assessment window. A timepoint is valid if it includes at least 5 valid nights, including at least 1 weekend night; otherwise it is treated as missing. Unit: minutes.

Change in 7-day mean sleep efficiency from baseline to Week 6

Change equals the Week 6 mean minus the baseline mean. Sleep efficiency is the percentage of time in bed scored as sleep for each valid night. Sleep is measured with Actiwatch 2 (Philips Respironics) worn on the nondominant wrist, recorded in 60-second epochs, and scored with Philips Actiware version 6.0.9 using sleep diaries. Each timepoint is the mean across all valid nights in the 7-day assessment window. A timepoint is valid if it includes at least 5 valid nights, including at least 1 weekend night; otherwise it is treated as missing. Unit: percentage points.

Change in Pittsburgh Sleep Quality Index total score over time

The Pittsburgh Sleep Quality Index total score ranges from 0 to 21; higher scores indicate poorer sleep quality. Change at Week 6 and at the optional approximately 3-month post-treatment follow-up equals the respective follow-up score minus baseline.

Change in 7-day mean number of awakenings from baseline to Week 6

Change equals the Week 6 mean minus the baseline mean. Actiware-derived awakenings are counted for each valid night. Sleep is measured with Actiwatch 2 (Philips Respironics) worn on the nondominant wrist, recorded in 60-second epochs, and scored with Philips Actiware version 6.0.9 using sleep diaries. Each timepoint is the mean across all valid nights in the 7-day assessment window. A timepoint is valid if it includes at least 5 valid nights, including at least 1 weekend night; otherwise it is treated as missing. Unit: awakenings per night.

Change in 7-day mean sleep fragmentation index from baseline to Week 6

Change equals the Week 6 mean minus the baseline mean. For each valid night, the Actiware 6.0.9 sleep fragmentation index is the sum of (1) the percentage of epochs classified as mobile during the sleep interval and (2) the percentage of all immobile bouts lasting no more than 1 minute. Higher values indicate greater sleep fragmentation. Actiwatch 2 is worn on the nondominant wrist and recorded in 60-second epochs. Each timepoint is the mean across valid nights in the 7-day window. A timepoint requires at least 5 valid nights, including at least 1 weekend night; otherwise it is missing. Unit: percentage points.

Change in CERAD-NB word-list memory composite score from baseline to Week 6

Change equals the Week 6 score minus baseline. The composite is the sum of word-list learning across 3 trials (0-30), delayed recall (0-10), and recognition discriminability (0-10), for a total range of 0-50; higher scores indicate better memory performance. Recognition discriminability is calculated as correct "Yes" responses plus correct "No" responses minus 10; values below 0 are scored as 0. Remote memory is not included in this composite.

Proportion of participants completing at least 10 of 12 acupuncture sessions

Number of enrolled participants who complete at least 10 of the 12 planned acupuncture sessions divided by the total number enrolled, multiplied by 100. This threshold corresponds to at least 80% treatment adherence. Unit: percentage of participants.

Proportion of participants completing the Week 6 primary outcome assessment

Number of enrolled participants who complete the Week 6 assessment of the primary outcome, 7-day mean wake after sleep onset, divided by the total number enrolled, multiplied by 100. Unit: percentage of participants.

Proportion of participants with valid actigraphy data at both baseline and Week 6

Number of enrolled participants with valid actigraphy timepoints at both baseline and Week 6 divided by the total number enrolled, multiplied by 100. Each timepoint requires at least 5 valid nights, including at least 1 weekend night. Unit: percentage of participants.